iflow-mcp_pingcy_app_chatppt 0.1.0__py3-none-any.whl

__init__.py

@@ -0,0 +1,3 @@
+"""App ChatPPT MCP Server - PowerPoint processing and chat capabilities."""
+
+__version__ = "0.1.0"

doubao.py

@@ -0,0 +1,58 @@
+import base64
+from openai import OpenAI
+from typing import List, Optional
+import os
+
+class DoubaoVisionLLM:
+    def __init__(
+        self, 
+        model_name: str,
+        api_key: Optional[str] = None,
+        base_url: str = "https://ark.cn-beijing.volces.com/api/v3"
+    ):
+        self.model_name = model_name
+        self.client = OpenAI(
+            base_url=base_url,
+            api_key=api_key or os.getenv("ARK_API_KEY")
+        )
+
+    def _get_image_mime_type(self, image_path: str) -> str:
+        """Get the MIME type based on the image file extension."""
+        ext = image_path.lower().split('.')[-1]
+        mime_types = {
+            'jpg': 'image/jpeg',
+            'jpeg': 'image/jpeg',
+            'png': 'image/png'
+        }
+        return mime_types.get(ext, 'image/jpeg')
+
+    def _image_to_base64(self, image_path: str) -> str:
+        with open(image_path, "rb") as image_file:
+            return base64.b64encode(image_file.read()).decode('utf-8')
+
+    def generate_response(
+        self,
+        prompt: str,
+        image_paths: List[str],
+        system_prompt: str = "你是一个结合文本与图片回答问题的AI助手。"
+    ) -> str:
+        completion = self.client.chat.completions.create(
+            model=self.model_name,
+            messages=[
+                {"role": "system", "content": system_prompt},
+                {"role": "user", 
+                 "content": [
+                    {"type": "text", "text": prompt},
+                    *[
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": image_path if image_path.startswith(("http://", "https://")) else f"data:{self._get_image_mime_type(image_path)};base64,{self._image_to_base64(image_path)}" ,
+                            },
+                        }
+                        for image_path in image_paths
+                    ]
+                ]},
+            ],
+        )
+        return completion.choices[0].message.content

iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/METADATA

@@ -0,0 +1,102 @@
+Metadata-Version: 2.4
+Name: iflow-mcp_pingcy_app_chatppt
+Version: 0.1.0
+Summary: Add your description here
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+Requires-Dist: asyncpg>=0.30.0
+Requires-Dist: browser-use>=0.1.24
+Requires-Dist: chroma>=0.2.0
+Requires-Dist: fastapi>=0.115.12
+Requires-Dist: langchain-core>=0.3.65
+Requires-Dist: langchain-mcp-adapters>=0.1.7
+Requires-Dist: langchain-openai>=0.3.14
+Requires-Dist: langgraph>=0.4.1
+Requires-Dist: llama-index-agent-openai>=0.4.6
+Requires-Dist: llama-index-core>=0.12.34.post1
+Requires-Dist: llama-index-embeddings-ollama>=0.6.0
+Requires-Dist: llama-index-embeddings-openai>=0.3.1
+Requires-Dist: llama-index-llms-openai>=0.3.38
+Requires-Dist: llama-index-tools-mcp>=0.1.2
+Requires-Dist: llama-index-vector-stores-chroma>=0.4.1
+Requires-Dist: mcp[cli]>=1.9.4
+Requires-Dist: openpyxl>=3.1.5
+Requires-Dist: pandas>=2.2.3
+Requires-Dist: pdf2image>=1.17.0
+Requires-Dist: pypdfium2>=4.30.1
+Requires-Dist: python-socks>=2.7.1
+Requires-Dist: redis>=6.1.0
+Requires-Dist: requests>=2.32.4
+Requires-Dist: streamlit>=1.45.0
+Requires-Dist: uvicorn>=0.34.3
+Requires-Dist: websockets>=15.0.1
+
+# ChatPPT-MCP:  多文档RAG引擎的 MCP Server
+
+一个基于MCP（Model Context Protocol）的多文档RAG（Retrieval-Augmented Generation）引擎应用，支持PPT文档的智能问答和分析。
+
+## 功能特性
+
+- 🔍 **多文档处理**: 支持索引多个PPT文档，基于视觉模型
+- 🤖 **智能问答**: 基于RAG技术的文档问答
+- 🔄 **MCP集成**: 使用Model Context Protocol进行工具调用
+- 📊 **交互式测试**: 提供命令行交互测试界面
+
+## 技术栈
+
+- **后端框架**: FastAPI
+- **向量数据库**: ChromaDB
+- **LLM**: Doubao Vision
+- **文档处理**: LibreOffice, pypdfium2
+- **协议**: Model Context Protocol (MCP)
+- **向量化**: OpenAI Embeddings
+
+## 安装使用
+
+### 1. 安装依赖
+
+```bash
+pip install -e .
+```
+
+### 2. 环境配置
+
+复制环境变量模板文件，并修改：
+
+```bash
+cp .env.example .env
+```
+
+### 3. 运行应用
+
+#### RAG引擎测试模式
+
+```bash
+python rag_interactive_test.py
+```
+
+#### MCP测试模式(先sse启动：python mcp_ppt_server.py --transport sse)
+
+```bash
+python mcp_interactive_test.py
+```
+
+## 项目结构
+
+```
+app_chatppt/
+├── src/                        # 源代码目录
+│   ├── multi_doc_rag_engine.py # RAG引擎核心
+│   ├── document_processor.py   # 文档处理器
+│   └── ...
+├── data/                       # 示例数据
+├── rag_interactive_test.py     # 交互式测试入口
+├── mcp_interactive_test.py     # MCP集成测试
+├── pyproject.toml              # 项目配置
+├── .env.example                # 环境变量模板
+└── README.md                   # 项目文档
+```
+
+## 许可证
+
+MIT License

iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+__init__.py,sha256=ieOvA37tdFuQDWzKhClZcBa3_T-YecF2IClZmjvrBq4,99
+doubao.py,sha256=DhGUfj9L5dAX67QporZlq5VYA8xd5B8Z913QTMZ56Ig,2044
+mcp_ppt_server.py,sha256=eM6q8DMnQ-k6kd-Np-R8Sc_ZbahXicNgjNz1lcAHw1o,8784
+multi_doc_rag_engine.py,sha256=pAtK6G20AcIyAyvHn4SvO-T6dYfzTp97TfVDhgw0YWs,40876
+ppt_utils.py,sha256=Jl4ldtkLA149botrkpfLf5dVMbR0PufVwIpvN1fGcjg,9408
+iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/METADATA,sha256=dYyDC933t89kMIjp1XYwmDVrkV4qTx-EahjyYWL9fKE,2834
+iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/entry_points.txt,sha256=8QHU_i81rOwO1TWnVMVXpSM5B030HRvKc_I8v-XNEjw,59
+iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/top_level.txt,sha256=KN7yztyd3Yq5ZZy_538BGbWffHA5I1Zleh_RQh8MpkA,62
+iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/RECORD,,

iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.10.2)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+mcp_ppt_server = src.mcp_ppt_server:main

iflow_mcp_pingcy_app_chatppt-0.1.0.dist-info/top_level.txt

@@ -0,0 +1,5 @@
+__init__
+doubao
+mcp_ppt_server
+multi_doc_rag_engine
+ppt_utils

mcp_ppt_server.py

@@ -0,0 +1,274 @@
+"""
+多文档PPT RAG引擎的MCP服务器
+"""
+
+import asyncio
+import json
+import os
+import sys
+from pathlib import Path
+from typing import Any, Dict, List, Optional, AsyncIterator
+import logging
+from contextlib import asynccontextmanager
+import argparse
+
+# MCP FastMCP导入
+from mcp.server.fastmcp.server import FastMCP, Context
+
+# 加载环境变量
+from dotenv import load_dotenv
+load_dotenv()
+
+# 本地导入
+from multi_doc_rag_engine import MultiDocRAGEngine
+
+# 配置日志
+logging.basicConfig(level=logging.INFO, 
+                    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+logger = logging.getLogger(__name__)
+
+
+# 定义应用程序上下文类
+class AppContext:
+    def __init__(self):
+        self.rag_engine: Optional[MultiDocRAGEngine] = None
+
+
+# 创建生命周期管理器
+@asynccontextmanager
+async def app_lifespan(server: FastMCP) -> AsyncIterator[AppContext]:
+    """管理应用程序生命周期，包括多文档RAG引擎初始化
+    
+    Args:
+        server: FastMCP服务器实例
+        
+    Returns:
+        应用程序上下文对象
+    """
+    # 启动初始化
+    logger.info("应用程序正在启动，初始化资源...")
+    app_ctx = AppContext()
+    
+    try:
+        # 从环境变量获取配置
+        cache_dir = os.getenv("CACHE_DIRECTORY", "./cache")
+        chroma_dir = os.getenv("CHROMA_PERSIST_DIRECTORY", "./chroma_db")
+        doubao_model = os.getenv("DOUBAO_MODEL", "ep-20250205153642-hzqpj")
+        
+        # 初始化多文档RAG引擎
+        app_ctx.rag_engine = MultiDocRAGEngine(
+            persist_dir=chroma_dir,
+            cache_dir=cache_dir,
+            doubao_model=doubao_model
+        )
+        
+        logger.info("多文档RAG引擎初始化成功")
+        
+        # 将上下文传递给应用程序
+        yield app_ctx
+        
+    finally:
+        # 关闭时清理资源
+        logger.info("应用程序正在关闭，清理资源...")
+
+
+# 创建MCP服务器实例
+app = FastMCP("multi-doc-ppt-rag", port=5053, lifespan=app_lifespan)
+
+
+@app.tool()
+async def add_ppt(
+    ctx: Context,
+    file_path: str,
+    force_reprocess: bool = False
+) -> str:
+    """将指定的PPT文档添加到RAG索引中
+    
+    Args:
+        ctx: 上下文对象
+        file_path: 要添加的PPT文件的绝对或相对路径
+        force_reprocess: 是否强制重新处理，即使文档已存在于索引中
+    
+    Returns:
+        操作结果的JSON字符串
+    """
+    try:
+        if not file_path:
+            return json.dumps({"status": "error", "message": "需要提供file_path参数"}, ensure_ascii=False)
+        
+        # 将相对路径转换为绝对路径
+        abs_path = str(Path(file_path).resolve())
+        if not os.path.exists(abs_path):
+            return json.dumps({"status": "error", "message": f"文件未找到: {abs_path}"}, ensure_ascii=False)
+        
+        # 从上下文获取RAG引擎
+        rag_engine = ctx.request_context.lifespan_context.rag_engine
+        
+        logger.info(f"开始添加PPT文档: {abs_path}")
+        await ctx.info(f"开始添加PPT文档: {abs_path}")
+        
+        result = await rag_engine.add_ppt_document(abs_path, force_reprocess=force_reprocess)
+        
+        logger.info(f"PPT文档添加完成: {result['status']}")
+        await ctx.info(f"PPT文档添加完成: {result['status']}")
+        
+        return json.dumps(result, indent=2, ensure_ascii=False)
+        
+    except Exception as e:
+        error_msg = f"添加PPT文档时出错: {str(e)}"
+        logger.error(error_msg)
+        return json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False)
+
+
+@app.tool()
+def delete_ppt(
+    ctx: Context,
+    file_path: str
+) -> str:
+    """从RAG索引中删除指定的PPT文档
+    
+    Args:
+        ctx: 上下文对象
+        file_path: 要删除的PPT文件的绝对或相对路径
+    
+    Returns:
+        操作结果的JSON字符串
+    """
+    try:
+        if not file_path:
+            return json.dumps({"status": "error", "message": "需要提供file_path参数"}, ensure_ascii=False)
+        
+        # 将相对路径转换为绝对路径
+        abs_path = str(Path(file_path).resolve())
+        
+        # 从上下文获取RAG引擎
+        rag_engine = ctx.request_context.lifespan_context.rag_engine
+        
+        logger.info(f"开始删除PPT文档: {abs_path}")
+        
+        if not rag_engine.is_document_indexed(abs_path):
+            return json.dumps({"status": "error", "message": f"文档不在索引中: {abs_path}"}, ensure_ascii=False)
+
+        result = rag_engine.remove_ppt_document(abs_path)
+        
+        logger.info(f"PPT文档删除完成: {result['status']}")
+        
+        return json.dumps(result, indent=2, ensure_ascii=False)
+        
+    except Exception as e:
+        error_msg = f"删除PPT文档时出错: {str(e)}"
+        logger.error(error_msg)
+        return json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False)
+
+
+@app.tool()
+async def chat_with_ppt(
+    ctx: Context,
+    query: str,
+    file_path: Optional[str] = None,
+    doc_id: Optional[str] = None
+) -> str:
+    """与一个或所有PPT文档进行对话
+    
+    Args:
+        ctx: 上下文对象
+        query: 用户提出的问题
+        file_path: (可选) 指定要查询的PPT文件路径。如果提供，则只在该文档中搜索
+        doc_id: (可选) 指定要查询的文档ID。如果提供，则只在该文档中搜索。如果同时提供了file_path和doc_id，则优先使用doc_id
+    
+    Returns:
+        查询结果的JSON字符串
+    """
+    try:
+        if not query:
+            return json.dumps({"status": "error", "message": "需要提供query参数"}, ensure_ascii=False)
+        
+        # 从上下文获取RAG引擎
+        rag_engine = ctx.request_context.lifespan_context.rag_engine
+        
+        logger.info(f"开始查询: {query}")
+        await ctx.info(f"开始查询: {query}")
+        
+        result = await rag_engine.query(query, file_path=file_path, doc_id=doc_id)
+        
+        logger.info(f"查询完成: {result['status']}")
+        await ctx.info(f"查询完成: {result['status']}")
+        
+        return json.dumps(result, indent=2, ensure_ascii=False)
+        
+    except Exception as e:
+        error_msg = f"查询时出错: {str(e)}"
+        logger.error(error_msg)
+        return json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False)
+
+
+@app.tool()
+def index_status(
+    ctx: Context
+) -> str:
+    """获取当前RAG索引的状态和统计信息
+    
+    Args:
+        ctx: 上下文对象
+    
+    Returns:
+        索引状态的JSON字符串
+    """
+    try:
+        # 从上下文获取RAG引擎
+        rag_engine = ctx.request_context.lifespan_context.rag_engine
+        
+        logger.info("获取索引状态")
+        
+        result = rag_engine.get_index_status()
+        
+        logger.info(f"索引状态: {result['status']}")
+        
+        return json.dumps(result, indent=2, ensure_ascii=False)
+        
+    except Exception as e:
+        error_msg = f"获取索引状态时出错: {str(e)}"
+        logger.error(error_msg)
+        return json.dumps({"status": "error", "message": error_msg}, ensure_ascii=False)
+
+
+def get_available_tools():
+    """获取当前服务器提供的所有工具"""
+    tools = app._tool_manager.list_tools()
+    return tools
+
+
+def main():
+    """MCP服务器的主入口点"""
+    # 命令行参数
+    parser = argparse.ArgumentParser(description="启动多文档PPT RAG MCP服务器")
+    parser.add_argument("--transport", choices=["sse", "stdio"], default="stdio", 
+                        help="选择传输方式: sse 或 stdio")
+    parser.add_argument("--port", type=int, default=5053, help="指定服务器端口号(仅用于SSE)")
+    args = parser.parse_args()
+    
+    # 更新端口设置
+    if args.transport == "sse":
+        current_settings = app.settings
+        current_settings.port = args.port
+        app.settings = current_settings
+    
+    # 列出可用工具
+    logger.info("\n")
+    logger.info("==== 多文档PPT RAG Server可用工具 ====")
+    tools = app._tool_manager.list_tools()
+    if tools:
+        for i, tool in enumerate(tools, 1):
+            description = tool.description.strip().split("\n")[0] if tool.description else "无描述"
+            logger.info(f"{i}. {tool.name}: {description}")
+    else:
+        logger.info("未发现工具函数")
+    logger.info("=================================\n")
+            
+    # 启动服务器
+    logger.info(f"启动多文档PPT RAG服务器，传输方式: {args.transport}, 端口: {args.port if args.transport == 'sse' else 'N/A'}")
+    app.run(transport=args.transport)
+
+
+if __name__ == "__main__":
+    main()